PUWF-AV: Pediatric Ultra-Widefield Fundus Images for Artery-Vein Segmentation

PUWF-AV is a dataset of 60 ultra-widefield (UWF) fundus images for retinal artery-vein segmentation. The images are from a pediatric population and were annotated by professional ophthalmologists. It was published by Chenkai Zhao on figshare under a CC-BY-4.0 license.

Use Cases

• Training segmentation models for retinal arteries and veins based on ultra-widefield fundus images.
• Developing methods for analyzing retinal vascular parameters in pediatric populations.
• Benchmarking artery-vein segmentation algorithms on a novel UWF imaging modality.

Strengths

• Contains 60 professionally annotated ultra-widefield fundus images.
• Focuses on a pediatric population, which the description notes is a gap in existing public data.
• Annotations were performed by professional ophthalmologists, suggesting clinical validity.

Limitations

• The dataset size of 60 images is relatively small for training complex deep learning models.
• Row and column-level metadata are unknown, requiring manual inspection after download.
• Geographic and demographic details of the pediatric cohort are not provided, which may limit generalizability.

Provenance

Source

figshare, author Chenkai Zhao.

Collection Method

Images were annotated for artery-vein segmentation by professional ophthalmologists.

Freshness

Last updated 2026-05-30 05:06:35.